We established polyclonal antibodies against the peptide sequences of proNT/NMN (amino acids (aa) 8–87) shown in Fig. 1A. Chemically synthesized proNT/NMN (aa 8–87) was conjugated with keyhole limpet hemocyanin (KLH) and then purified. Antibodies were obtained by immunizing three SPF Japanese white rabbits with KLH-conjugated peptides.

SBC3, a small cell lung carcinoma cell line, was purchased from the Japanese Cancer Research Bank. SBC3 cells were maintained in RPMI-1640 (Sigma, St. Louis, MO, USA) medium supplemented with 10% FBS (Gibco, Grand Island, NY, USA), penicillin (100 U/ml), and streptomycin (0.1 mg/ml) in a humidified 5% CO₂ incubator. Serum-free conditioned media was prepared as previously described (14).

Six-week-old athymic nude mice (BALB/cAJcl- nu/nu) were purchased from CLEA Japan (Tokyo, Japan), and NRG mice (NOD.Cg-Rag1^tm1MomIl2rg^tm1Wjl/SzJ) were purchased from Charles River (Wilmington, MA, USA). NRG mice were bred in the Experimental Animal Facility of Shizuoka Cancer Center Research Institute, and their offspring was used for experiments. All animal experiments were carried out in accordance with the general guidelines for animal experiments of Shizuoka Cancer Center Research Institute.

Nude mice were subcutaneously inoculated with SBC3 cells (5×10⁶ cells in 0.1 ml PBS/mouse) into the right forelimb, and blood samples were collected once a week. The tumor size was also measured before blood collection to calculate the tumor volume. NRG mice were intradermally inoculated with SBC3 cells (5×10⁶ cells in 0.1 ml PBS/mouse) to examine the effect of tumor resection, and divided into two groups (n=5 per group). Four weeks after inoculation, tumors were surgically resected from one group and blood was collected from the non-resected and control mice after 24, 48 and 120 h. Blood samles were centrifuged, and plasma was collected. Plasma samples were stored at −80°C until use.

Frozen tumor tissues were treated with lysis buffer containing 7.5 M urea, 2.5 M thiourea, 12.5% glycerol, 50 mM Tris-HCl (pH 7.4), 2.5% N-octylglucoside, 6.25 mM Tris-carboxyethyl phosphine hydrocholine (TCEP), and 1.25 mM protease inhibitor, and homogenized by TissueLyser II (Qiagen K.K., Tokyo, Japan). Following rotation for 1 h, the samples were centrifuged at 14,000 × g for 30 min at 4°C, and the supernatants were collected. Protein concentrations were determined by the Bradford assay (Bio-Rad Laboratories, Hercules, CA, USA).

Microtiter plates (96-well, Nunc C8 Maxisorp Immuno module; Nunc, Roskilde, Denmark) were coated with 100 μl of goat anti-rabbit IgG (Fc) solution (diluted 1:200 in 0.1 M NaHCO₃; MP Biomedicals, Irvine, CA, USA) and incubated for 24 h at 4°C. After incubation, the antibody solution was removed, and the wells were blocked overnight at 4°C by using block ace solution (DS Pharma Biomedical Co., Ltd., Osaka, Japan). Following three washes with 0.05% Tween-20 in PBS, the biotinylated antigen solution [20 ng/ml biotinyl-proNT/NMN (aa 40–87), prepared in PBS containing 0.1% BSA], the plasma samples, and the antibody solution [anti-proNT/NMN rabbit polyclonal antibody (105S), 1/3,000 dilution] were added and incubated with shaking for 3 h at room temperature. Plates were washed with PBS containing 0.05% Tween-20 four times before the addition of labeling solution (streptavidin, horseradish peroxidase conjugate; Merck, Darmstadt, Germany), and incubated for 2 h at room temperature. After four final washes, BM chemiluminescence ELISA substrate (POD) solution (Roche Applied Science, Mannheim, Germany) was added to each well and allowed to develop in the dark for 3 min. The chemiluminescence of BM ELISA was measured using Wallac 1420 ARVO MX (Perkin-Elmer Life Sciences, Waltham, MA, USA). A standard curve for proNT/NMN levels was obtained using proNT/NMN (aa 40–87) as the standard antigen.

Proteins were separated by SDS-PAGE on 14.25% polyacrylamide gels and electrophoretically transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). Immunoblotting was performed using anti-proNT/NMN rabbit polyclonal antibodies (105S, 1:1,000 dilution), or anti-neurotensin goat polyclonal antibodies (N-17 and C-19, 1:1,000 dilution; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) as the primary antibody, and anti-rabbit IgG-, or anti-goat IgG-horseradish peroxidase (HRP) conjugate (1:10,000 dilution; Jackson ImmunoResearch, West Grove, PA, USA) as the secondary antibody. HRP-dependent luminescence was developed using ECL Prime Western blotting detection reagent (GE Healthcare UK Ltd., Buckinghamshire, UK) and detected in a LAS-3000 (Fuji Film, Tokyo, Japan).

Formalin-fixed tumor tissues were gradually dehydrated in ethanol and embedded in paraffin. Tissue sections (3-μm thick) were cut and subjected to immunohistochemistry. Sections were deparaffinized in xylene, and treated with 1% H₂O₂ in methanol for 30 min to block the endogenous peroxidase activity. After blocking, they were incubated with anti-proNT/NMN polyclonal antibody (1:200 dilution) or with the antibody in the presence of the antigen peptide for 30 min at room temperature. The sections were then washed with PBS and incubated with peroxidase labeled polymer (goat anti-rabbit IgG; Dako, Glostrup, Denmark) for 30 min. Following one wash with PBS, peroxidase activity was initiated with DAB substrate solution (Dako). The peroxidase reaction was completed, the sections were washed with PBS, counterstained with hematoxylin, dehydrated, cleared in xylene, and a coverslip was placed over the sections.

The results of the experiments are expressed as the mean ± SE values. The statistical significance of differences was determined according to the Student’s t-test.

We first established a rabbit polyclonal antibody against the proNT/NMN (aa 8–87) and developed a competitive ELISA system. Based on the results of the standard curve with proNT/NMN (aa 40–87) (Fig. 1B), the proNT/NMN levels in mouse plasma were measured by ELISA. Every week, plasma samples were collected from the control and SBC3-inoculated mice after inoculation. As shown in Fig. 2A, tumor volumes of SBC3-inoculated mouse gradually increased. Correlating with tumor volumes, the plasma proNT/NMN levels of the tumor-bearing mice were elevated, while the proNT/NMN levels in control mouse were not significantly detectable.

To further investigate proNT/NMN secretion, tumors were surgically resected from the NRG mice intradermally inoculated with SBC3 cells, and the plasma proNT/NMN levels were evaluated after resection. Compared to nude mice that were subcutaneously inoculated with SBC3 cells, the tumors of NRG mice grew slowly, and the plasma proNT/NMN levels were slightly elevated 4 weeks after inoculation (nude mice, 128.49±32.18 pmol/ml plasma; NRG mice, 22.93±4.67 pmol/ml plasma). The differences in the plasma proNT/NMN levels between the control and intradermally inoculated NRG mice were statistically significant. After surgical resection, the plasma proNT/NMN levels of the tumor-resected mice decreased until they reached the level identical to that of the control mice (Fig. 2B). Although the levels of the tumor-bearing mice (non-resected mice) were significantly elevated after 120 h, the levels of the tumor-resected mice remained the same as that of the control mice. These results clearly suggest that SBC3 tumors secreted proNT/NMN into blood.

Plasma proNT/NMN was also detected by immunoblotting using the same antibody as used in the ELISA. In SBC3 serum-free conditioned medium, the positive band was detected around 17 kDa (Fig. 3A left line, SBC3-CM). The medium contains proNT/NMN (14), and the theoretical molecular mass of proNT/NMN is about 17 kDa, suggesting that the band detected around 17 kDa is specific to proNT/NMN. In mouse plasma, several bands were detected around the molecular mass of 17 kDa. The band at 17 kDa, specific to full-length proNT/NMN, was only detected in tumor-bearing mouse plasma; however, the lower bands (~14–16 kDa) were detected in the plasma samples from both the tumor-bearing and control mice (Fig. 3A).

To gain more insight into whether the lower bands are derived from proNT/NMN, immunoblotting was performed using anti-NT antibodies, N-17 and C-19. Both antibodies can recognize NT-containing products such as NT, large NT, proNT/NMN, and N-17 can also recognize large NMN. Using these two antibodies, the bands around 14–16 kDa were detected in the plasma of tumor-bearing mouse, but not in the plasma from the control mouse (Fig. 3B). These results suggest that the lower band detected in the control mouse plasma is not derived from secreted proNT/NMN, but in the tumor-bearing mouse plasma, the processing products of proNT/NMN such as large NMN and large NT, were detected around 14–16 kDa.

ProNT/NMN expression in mouse tumor tissues was also determined by immunoblotting and immunohistochemistry. As with the plasma samples, the positive bands derived from proNT/NMN were detected in the SBC3 tumor tissues (Fig. 4A). In addition, Fig. 4B shows that proNT/NMN was expressed particularly in the superficial layer of the tumor (left panel). Immunostaining was diminished when the tumor tissue section was treated with antibody in the presence of the antigen peptide (right panel). This peptide competition analysis confirmed the expression of proNT/NMN in the above-mentioned areas in the tumor tissue. ProNT/NMN expression in the tumor tissues indicated that this precursor was produced by SBC3 tumors and secreted into plasma as the tumors grew.